Blood Glucose Control Based On TS Fuzzy Model

Diabetes mellitus is due to genetic factors,immune dysfunction,microbial in-fection,mental factors,and so on.The loss of insulin secretion in the human body or insulin resistance will break the balance of blood glucose and result in a disease that persists in hyperglycemia.With the improvement of the living of material and the deterioration of the environment,Diabetes mellitus has become a multi-ple disease which seriously endangers human health.At present,there is no way to cure diabetes mellitus.Artificial pancreas is produced to keep the balance of blood glucose.Designing and testing algorithms based on blood glucose closed loop control system is the key of artificial pancreas.Since the chemical reaction process of the glucose-insulin system is nonlinear,traditional control theory is difficult to model and control such complex system.Therefore,it is important to control the glucose-insulin system model.Because the Takagi-Sugeno(T-S)fuzzy model can approximate a smooth nonlinear function with arbitrary precision,so linear control theory methods can be applied to analyze and control nonlinear systems.Therefore,the T-S fuzzy method is one of the effective means to study glucose-insulin control system.The purpose of this paper is to approximate the glucose-insulin control system model by T-S fuzzy model.Then,based on Lyapunov-Krasovskii stability theory,and we apply method of parallel distribution compensation and linear matrix in-equaty technique to study the H_?control problem of glucose-insulin system,which will be useful to control the blood glucose of diabetes.First,the structure and working mechanism of the glucose-insulin system and artificial pancreas are introduced.Secondly,the H_? control problem of linear T-S fuzzy systems.By constructing a suitable Lyapunov-Krasovskii function,and using parallel distributed compensa-tion method,a feedback controller with certain constraints is designed.The global asymptotic stability of the system represented by a set of linear matrix inequalities,and the H_?performance is simultaneously satisfied.Finally,the glucose-insulin control system model is approximated as the T-S fuzzy model.The blood glucose controller with constraints is designed by applying the obtained global asymptotic stability condition.By using the MATLAB toolbox,simulation is carried out to verify the validity of the theory.